Revealing cell-level gene regulation through integration of single-cell multi-omics measurements

通过整合单细胞多组学测量揭示细胞水平的基因调控

• 批准号: 10710022
• 负责人: Yijie Wang
• 金额: $ 38.05万
• 依托单位: TRUSTEES OF INDIANA UNIVERSITY
• 依托单位国家: 美国
• 财政年份: 2022
• 资助国家: 美国
• 起止时间: 2022-09-26 至 2027-07-31
• 项目状态: 未结题
• 来源: https://reporter.nih.gov/project-details/10710022
• 关键词: Cells; Computing Methodologies; Distal; Foundations; Gene Expression Regulation; Measurement; Organism; Post-Transcriptional Regulation; Prevention; Regulatory Element; Research; Resolution; Solid; Techniques; Transcriptional Regulation; cell type; computer framework; disease diagnosis; insight; machine learning algorithm; multiple omics; single cell sequencing

Summary Advanced single-cell sequencing techniques have enabled us to infer gene regulation at the single-cell level. We propose to develop computational methods to overcome obstacles for elucidating gene regulation at single-cell resolution. We first present an alignment-based computational framework to integrate single-cell multi-omics measurements. The alignment-based computational framework can effectively handle the cell type imbalance problem and is more robust to hyperparameters. Furthermore, we incorporate the integrated single-cell multi-omics measurements and advanced machine learning algorithms to infer transcriptional regulation, distal regulatory elements, and post-transcriptional regulation at the single-cell level. We expect to develop computational methods to better understand gene regulation, which would lay a solid foundation for disease diagnosis, treatment, and prevention.

总结 先进的单细胞测序技术使我们能够在单细胞水平上推断基因调控。我们 建议开发计算方法，以克服阐明单细胞基因调控的障碍 分辨率我们首先提出了一个基于单细胞多组学的计算框架， 测量.基于网格的计算框架可以有效地处理单元类型不平衡 并且对超参数更鲁棒。此外，我们将整合的单细胞多组学 测量和先进的机器学习算法来推断转录调控，远端调控， 元件，和转录后调节在单细胞水平。我们希望发展计算方法 更好地了解基因调控，为疾病诊断、治疗和治疗奠定坚实的基础 预防